JOURNAL OF PLANKTON RESEARCH j VOLUME 30 j NUMBER 1 j PAGES 57 – 64 j 2008

Concentration of fixed plankton samples

via settling: how long is long enough?
MONIKA CLAESSENS* AND MARIO PRAST
DEPARTMENT OF ORGANISMAL BIOLOGY, UNIVERSITY OF SALZBURG, HELLBRUNNER STRASSE 34, 5020 SALZBURG, AUSTRIA

*CORRESPONDING AUTHOR: monika.claessens@sbg.ac.at

Downloaded from https://academic.oup.com/plankt/article-abstract/30/1/57/1494425 by guest on 10 April 2019

Received July 24, 2007; accepted in principle October 31, 2007; accepted for publication November 20, 2007; published online
November 23, 2007

Communicating editor: K.J. Flynn

Enumerating plankton cells, especially ciliates, usually requires settling of samples in order to con-
centrate the cells. Poorly settled samples could introduce large errors into plankton counts. The time
sufficient to settle all ciliates has, however, never been established in the literature. Here, using both
theoretical and empirical studies, we suggest improvements of the current method, which mostly
relies on experience to determine settling times. Ciliate density was used to calculate the theoretical
settling time of fixed ciliates with the Stokes equation. To determine ciliate density (g mL21), we
modified and established a density gradient centrifugation method. We found that ciliate density
was in the range 1.02 – 1.08 g mL21. Additionally, empirical sinking rates were gathered semi-
automatically with a digital camera system. The theoretical and experimental settling times were
in the same range, though there were differences for some species. From this, we recommend
working with the empirical sinking rates that are more reliable: 0.5 and 1.7 mm min21 for fixed
marine (at salinities of 16 and 40, respectively) samples and 2.4 mm min21 for fixed freshwater
samples. Using these rates potentially saves up to 95% of the time for settling compared to old,
experience derived times. Although ciliate density was significantly correlated with settling rates,
there was no correlation with particle size and shape.

I N T RO D U C T I O N empirical sinking velocities for a wide range of cell

types and shapes and culture salinities (viscosities).
The sinking velocities of unfixed cells or particles have
Empirical sinking velocities were gathered with a semi-
been addressed in many studies, mostly regarding phy-
automatic digital camera system. We expected to find (i)
toplankton, marine snow or other particles (e.g. Hamm,
an increase in sinking velocity with an increase in cell
2002; Kiorboe et al., 2002, 2003; Ptacnik et al., 2003;
size/diameter, (ii) that differences in cell shape com-
Peterson et al., 2005). The Utermöhl method
pared with spherical forms affect the sinking velocity
(Utermöhl, 1958) has been historically used for fixed
and finally (iii) that marine samples will settle more
plankton, but data regarding the duration of settling of
slowly than freshwater samples due to the increased
the fixed samples rely mostly on experience or on esti-
density of seawater.
mation. For phytoplankton, at least few studies were
conducted to achieve reliable sinking velocities (Padisák
et al., 2003 and literature cited therein). The Utermöhl
method is regularly used for ciliates, but as far as we
know, there are no data available for the optimal
METHODS
settling time, though this is the critical point for obtain-
ing reliable abundance data. Ciliate cultures
Here, we compare experimentally collected data We used seven ciliate cultures, five freshwater cultures
about sinking velocities with theoretical data, following and two from marine systems. Tetrahymena pyriformis and
the Stokes equation. We determine both theoretical and Colpidium colpoda were isolated from the River Salzach

doi:10.1093/plankt/fbm095, available online at www.plankt.oxfordjournals.org

# The Author 2007. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org
 JOURNAL OF PLANKTON RESEARCH j VOLUME
(Austria), Cyclidium glaucoma was provided by the
University Bielefeld (M. Bergtold; Germany), Paramecium
aurelia was from Culture Collection of Algae and
Protozoa (CCAP, UK), Euplotes octocarinatus was provided
from the University Stuttgart (H.-D. Görtz; Germany).
The marine cultures were multispecies assemblages of
different size and cell shape which were grown from
natural, prefiltered (100mm) seawater. Marine ciliates
are extremely sensitive to changes in the environment
and therefore we could not obtain single species cul-
tures. The Baltic Sea cultures were isolated from the
Bay of Kiel (Germany) and the Red Sea ciliates were
isolates from the Gulf of Aqaba. Freshwater ciliate
stocks were cultured in a 1:1 mixture of Volvic mineral
water and SMB-medium (1.5 mM NaCl, 0.05 mM
KCl, 0.4 mM CaCl2 . 2H2O, 0.05 mM MgCl2 . 6H2O,
0.05 mM MgSO4 . 7H2O, 2.0 mM, sodium phosphate
buffer, pH 6.8). Ciliates from the Baltic Sea and the
Red Sea were cultured in artificial seawater with sali-
nities of 16 and 40, respectively. Size and shape of the
ciliates were examined with a Nikon Eclipse E-800
microscope, equipped with a Nikon DXM1200F digital
camera supported by LuciaNet software, at 400 –
1000 magnification.
Determination of ciliate density
Ciliate density was determined with a density gradient
centrifugation (Fig. 1). Ciliates were fed with black
Indian ink for 30 min to increase visibility of the ciliate
band after the density gradient centrifugation. The
ingested ink particles were not expected to change cell
density. Thereafter, the ciliates were fixed with glutar-
dialdehyde (2% f.c.). Glutardialdehyde was chosen
over another commonly used fixative (e.g. in Bouin’s:
Fig. 1. The procedure after the density gradient centrifugation; the
distances between the water surface and the bead bands were
measured and used to create the standard curves. With the standard
curves and the distance from surface to the ciliate bands, the ciliate
densities were calculated.
58
30 j NUMBER 1 j PAGES 57 – 64 j 2008
formaldehyde saturated with picric acid with 10%
glacial acetic acid), because the latter often leads to pre-
cipitation of the centrifugation media, Percoll. Cell
shrinkage is a well-known phenomenon after fixation,
with different degrees of shrinkage depending on the
used fixative (Leakey et al., 1994; Stoecker et al., 1994)
which might increase cell density and thus affect the
settling velocity. The cell shrinkage is lower with glutar-
dialdehyde than with Bouin’s solution; the latter is
widely used for fixation of Bouin’s (Leakey et al., 1994;
Stoecker et al., 1994). Thus the settling velocity in Boui’s
Downloaded from https://academic.oup.com/plankt/article-abstract/30/1/57/1494425 by guest on 10 April 2019
samples should be even higher than the values Bouin’s
observed with glutardialdehyde due to the expected
higher density. After fixation, the cultures were concen-
trated via centrifugation (Heraeus Multifuge 4 KR, 61g,
10 min). The density gradient was developed for 30 min
with 20 256g (Sigma Percoll, autoclaved; Sorvall centri-
fuge RC-5Bþ, SLA-1500 rotor, Nunc 10 mL Oak
Ridge centrifuge tubes). One tube was loaded with dif-
ferentially coloured density marker beads (Amersham
Biosciences, 1.018, 1.033, 1.049, 1.062, 1.075 g mL21,
15 mL of each beadþ1925 mL sterile SMB-Medium)
and the remaining gradients were loaded with 2 mL of
the concentrated and fixed ciliate cultures (4 – 9 repli-
cates per species). Gradients were then centrifuged for
25 min with 385g (Sorvall RC-5Bþ, SLA-1500 rotor).
The distance from the water/Percoll interface to the
bands (beads or ciliates) was measured with a caliper
gauge (Fig. 1). Ciliate densities were calculated as the
difference of the ciliate band from the media surface
relative to that of bead bands of known density. For
each species, densities were calculated with the help of
standard curves, derived from the marker bead data
(Fig. 2; Table I).
Empirical sinking velocities of fixed ciliates
Ciliates from working stocks were fixed with glutardial-
dehyde (2% f.c.) and transferred into a sedimentation
chamber (Hydrobios, Ø = 2.56 mm, h = 18.5 mm, v =
9.8 mL) (Fig. 3). The chamber was immediately placed
on an inverted microscope (Nikon Eclipse TE2000-U)
and picture sequences were taken automatically using a
Nikon Digital Sight DS-SM camera and DS-L1
imaging computer. The intervals between two pictures
were 15– 30 s (3 – 4 replicates). Ciliates on the bottom of
the chamber were counted for each picture until the
number no longer increased (Fig. 3). The sinking vel-
ocity was calculated as this final sinking time and height
of the chamber. We never observed effects of convec-
tion. The final number of ciliate cells on the bottom of
the chamber ranged from 3 to 267 cells, depending on
the abundance in the culture that was used.
 M. CLAESSENS AND M. PRAST j SETTLING OF FIXED PLANKTON SAMPLES

Downloaded from https://academic.oup.com/plankt/article-abstract/30/1/57/1494425 by guest on 10 April 2019

Fig. 2. Calibration curves for ciliate density of the freshwater (top) and marine cultures used in the study (bottom). The x-axis shows the
distances, measured from water surface to the different bead bands; the y-axis shows the density, g mL21. The calibration curves were made with
the density marker beads data for each culture and replicate separately. (1) Paramecium aurelia; (2) Tetrahymena pyriformis; (3) Colpidium colpoda; (4)
Euplotes octocarinatus; (5) Cyclidium glaucoma.

Calculation of sinking velocity: Stokes

equation The sinking velocity (V; mm min21) was calculated
Sinking velocities were calculated following a modifi- with the following parameters (1): g, acceleration of
cation of the Stokes Law (Vogel, 1983) to allow a com- gravity (m s22); r, radius of the particle (mm); d, density
parison between empirical and theoretical settling of the sinking particle (g mL21); d1, density of the
velocities. medium (g mL21); F, form resistance; m, viscosity of the
medium (kg m22 s21).
To calculate form resistance, we had to develop an
2 d  d1 equation, which did not include velocity as a parameter;
V ¼ g r2 ð1Þ otherwise, we would have had to use our empirical
9 mF

59
 JOURNAL OF PLANKTON RESEARCH j VOLUME 30 j NUMBER 1 j PAGES 57 – 64 j 2008

Table I: Equations and R2 of the calibration

curves, which were used to calculate ciliate a, b and c are semi-axes of the cell (mm; half-axis of
21
densities (f(x) gives the cell density g mL ; x length, width and height).
is the distance mm) sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Species Equation of calibration curve R2 2 2 d  d1
F¼ gr K ð3pmDÞ ð3Þ
Cyclidium glaucoma f(x) = 3.442  1022x þ 1.018 0.95
9 m
Tetrahymena pyriformis f(x) = 3.60  1022x þ 9.953  1021 0.79
Paramecium aurelia f(x) = 9.508  1022x þ 9.296  1021 0.95
Euplotes octocarinatus f(x) = 2.44  1022x þ 1.017 0.94 pffiffiffiffiffiffiffiffi
1 a2 c2
Colpidium colpoda f(x) = 4.395  1022x þ 9.997  1021 0.99 a ¼ pffiffiffiffiffiffiffiffi
2
2 2
tanh a
Red Sea f(x) = 9.015  1022x þ 9.998  1021 0.89 a c pffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffi ð4Þ

Downloaded from https://academic.oup.com/plankt/article-abstract/30/1/57/1494425 by guest on 10 April 2019

2a2 1 a2 c2 a2 c2
Baltic Sea f(x) = 1.144  1021x þ 1.004 0.97 b ¼ 2 2 3=2 tanh a  a
ða c Þ
Calibration curves were made with the marker bead data.

a . b = c; a is the axis in sinking direction; calculates

shortest possible settling time due to the orientation of
the cell.
pffiffiffiffiffiffiffiffiffi
1 b2 a2
a ¼ pffiffiffiffiffiffiffiffi
2 ffi
2 2
tanh b
b a
pffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffi ð5Þ
a2 b b2 a2
 tanh1 b ba
2 2
b ¼ 2 2 3=2 a2
ðb a Þ

a = c , b; a is the axis parallel to the sinking direc-

tion, indicating that for equation (5) the longest axis b is
horizontal to the settling direction, increasing the form
resistance. Therefore, equation (5) calculates the longest
possible settling.
Equations (4) and (5) were used for T. pyriformis, C.
colpoda, P. aurelia and C. glaucoma. Theoretically, there are
two extreme possibilities for the orientation of the cell
regarding the moving direction: (i) when the long axis is
orientated parallel to the sinking direction, resulting in
Fig. 3. Exemplary data for the empirical determination of settling the shortest possible settling time, (4) and (6); and (ii) in
velocity. The x-axis shows the time period (seconds) in which pictures the case where the long axis is orientated perpendicular
where automatically taken. The y-axis shows the number of already
settled cells.
to the sinking direction, resulting in the longest possible
settling time, equations (5) and (7). Owing to the special
shape of Euplotes, for this species, the following
equations (6) and (7) were used (for details see McNown
settling velocities. For that we solved equation (2) and (1)
and Malaika, 1950).
for V and equalled them to get equation (3). Equation (2)
and those for a and b [equations (4) and (5)] were
derived from McNown and Malaika (McNown and 2F ðd1 ; K1 Þ 2a2 ½F ðd1 ; K1 Þ  Eðd1 ; K1 Þ
a ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi b¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Malaika, 1950). a and b were needed to calculate K (a a2  c2 ða2  b2 Þ a2  c2
proportionality factor) and with that form resistance cor- pffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffi
rectly, considering the cell shapes of the species used in a2  c2 a2  b2
d1 ¼ sin1 k1 ¼
our study [same for equation (6) and (7)]. a a2  c2
ð6Þ

F ¼ ð3pmVDÞK
a . b . c; a is the axis parallel to the sinking direc-
16 ð2Þ
1=3 tion and for equation (6) a is the longest axis. Therefore,
D ¼ 2ðabcÞ ; K ¼ Dða þ bÞ
3 the form resistance is lowest and equation (6) calculates

60
 M. CLAESSENS AND M. PRAST
the shortest possible settling time.
 
2F ðd2 ; K2 Þ 2a2 c Eðd2 ; K2 Þ
a ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffi b¼ 2  pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
b2  a2 c  a2 ab b2  a2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 2 b2  c2
1 b  a
d2 ¼ sin k2 ¼
b b2  a2
b . c . a; a is the axis in sinking direction; calculates
longest possible settling time.
For equations (6) and (7), E and F are elliptic integrals,
which were solved with tabulated values from Bronstein
and Semendjajew (1974). In addition to the ciliate species
used in the experiments, sinking velocities were also cal-
culated for Strombidium epidemum (10 mm, Claessens et al.,
2007) and Pelagodileptus tracheloides (up to 800 mm, Foissner,
1999), using a mean density of 1.05 g mL 21, to include
very small and large ciliate species.
R E S U LT S
Comparison of the empirical sinking rates with the
theoretical data indicates some effects of body shape on
the sinking velocities which could not be well modelled
by the modified Stokes equation. Therefore, we
Fig. 4. Box-plot of the ciliate densities (g mL21). The two marine
assemblages are on the left side, whereas the freshwater species are
arranged with increasing size on the right. BS, Baltic Sea; RS, Red
Sea, CG, Cyclidium glaucoma, TP, Tetrahymena pyriformis, EO, Euplotes
octocarinatus, CC, Colpidium colpoda, PA, Paramecium aurelia. The top,
bottom and line through the middle of the box correspond to the
75th percentile, 25th percentile and 50th percentile (median),
respectively. The whiskers on the bottom extend the 10th percentile
(bottom decile) and top 90th percentile (top decile).
j SETTLING OF FIXED PLANKTON SAMPLES
attribute great importance to the comparison of two
different methods to produce reliable sinking velocities.
Ciliate density
ð7Þ
Ciliate densities determined for the freshwater cultures
were not different from those of the marine cultures.
The freshwater species were between 1.02 and
Downloaded from https://academic.oup.com/plankt/article-abstract/30/1/57/1494425 by guest on 10 April 2019
1.08 g mL21, whereas the Red Sea and Baltic Sea cili-
ates had densities of 1.05 and 1.08 g mL21, respectively
(Fig. 4; Table II). The lowest densities were found for
T. pyriformis and C. colpoda (1.02 g mL21 for both), which
were significantly lower than the highest density
(P. aurelia and the Baltic Sea ciliates, 1.08 g mL21 for
both; ANOVA P , 0.001). Because of these differences,
we did not use a mean ciliate density for the calculation
of the settling time for the species used in our exper-
iments, but calculated the velocity for each species on
its own. For S. epidemum and P. tracheloides, a mean
density was used because no cultures were available.
Sinking velocity of fixed ciliates
Sinking velocities were experimentally determined for
the five different freshwater species and the two marine
assemblages (Table II). Ciliates differed in cell shape
and size, from the small marine species (19 mm length)
to P. aurelia (140 mm; Table II). The empirical sinking
velocities varied between 2.4 and 10.3 mm min21 for
the freshwater species and between 0.5 and
2.7 mm min21 for the ciliate assemblages from the Red
Sea. The lowest sinking rates were found for the Baltic
Sea ciliates (0.5 and 0.6 mm min21; Fig. 5). The settling
rates of both marine ciliate assemblages were signifi-
cantly lower than sinking velocities of the freshwater
species (Baltic Sea versus freshwater: P = 0.002; Red Sea
versus freshwater: P = 0.039). Though there was a sig-
nificant correlation between the empirical velocity and
ciliate density (P = 0.034; Pearson’s r = 0.96), there were
no correlations between ciliate size and sinking velocity
(P = 0.285; Pearson’s r = 0.24).
Sinking velocities were calculated for the freshwater
species only, as the cell size in the marine assemblages
was too diverse. The calculated velocities were between
2.1 and 5.3 mm min21, in the same range as the
empirical values for the same species (Table II; Fig. 5).
For C. glaucoma and C. colpoda, the calculated sinking vel-
ocities were clearly lower than the experimental velocity.
61
 JOURNAL OF PLANKTON RESEARCH j VOLUME 30 j NUMBER 1 j PAGES 57 – 64 j 2008

Table II: Characteristics of the ciliate cultures, used for the studies
Size Calculated sinking velocity Experimental sinking velocity Ciliate density
Ciliates (mm) Cell shape Salinity (mm min21) (mm min21) (g mL21)

Cyclidium glaucoma 23 Spherical FW 2.1 9.4 (4) 1.04 (6)

Tetrahymena 60 Ovoid FW 2.4 2.4 (4) 1.02 (8)
pyriformis
Paramecium aurelia 140 Ovoid –flattened FW 5.3 4.5 (3) 1.08 (4)
Euplotes octocarinatus 100 Flattened FW 4.4 10.3 (4) 1.05 (9)
Colpidium colpoda 110 Ovoid with nose FW 2.8 7.7 (4) 1.02 (9)
Baltic Sea Large 38 Mostly ovoid 16 n.d. 0.6 (4) 1.08 (6)
Baltic Sea Small 19 Mostly ovoid 16 n.d. 0.5 (4) 1.08 (6)
Red Sea Small 20 Ovoid and 40 n.d. 1.7 (3) 1.05 (8)

Downloaded from https://academic.oup.com/plankt/article-abstract/30/1/57/1494425 by guest on 10 April 2019

flattend
Red Sea Large 52 Ovoid and 40 n.d. 2.7 (3) 1.05 (8)
flattend
Strombidium 10 Ovoid 40 6.3 n.d. n.d.
epidemum
Pelagodileptus 800 Ovoid with FW 11.6 n.d. n.d.
tracheloides proboscis

Calculated sinking velocity is based on Stokes Equation (1) and is compared to the experimentally determined velocity. The salinities for the marine
cultures are given, whereas FW indicates freshwater cultures. Number of replicates is given in parentheses.

DISCUSSION (iii) marine samples have lower settling velocities due to

We had three hypotheses regarding the settling of fixed
ciliate samples: (i) settling velocities increase with
increasing cell size/diameter, (ii) variations from the
spherical body shape causes increased sinking times and
Fig. 5. Empirical (box-plots) and calculated (circles) sinking velocities
arranged with increasing ciliate size from left to right. The calculated
settling rates are given for the freshwater ciliates, but not for the
marine cultures due to the diversity in the size structure. BS, Baltic
Sea; RS, Red Sea; CG, Cyclidium glaucoma; TP, Tetrahymena pyriformis;
EO, Euplotes octocarinatus; CC, Colpidium colpoda; PA, Paramecium aurelia;
S, small; L, large. The top, bottom and line through the middle of the
box correspond to the 75th percentile, 25th percentile and 50th
percentile (median), respectively. The whiskers on the bottom extend
the 10th percentile (bottom decile) and top 90th percentile (top
decile).
the higher density of seawater compared with
freshwater.
Sinking velocities were not correlated with cell size,
but there was a significant correlation with ciliate
density. Cell shape was less important for the sinking
velocities as the two species having the highest sinking
velocities (E. octocarinatus 7.7 mm min21, C. glaucoma
9.4 mm min21) differed strongly in cell size and cell
shape (Table I). Euplotes is dorsoventrally flattened with
a band of cilia, the membranelle and several clusters
of tightly arranged cilia (cirri). These morphological
characteristics were expected to increase form resist-
ance relative to more spherical cells (e.g. Cyclidium).
The prominent membranelles and cirri at the edge of
the cell might cause the high variance within the
replicates of E. octocarinatus. These membranelles may
strongly affect the orientation of the cell and therefore
increase the settling time compared with spheroid
shaped cells. This became clear when comparing
minimal and maximal settling times of Cyclidium and
Euplotes, 8.3– 10.7 and 5 – 75 mm min21, respectively.
Additionally, the calculated settling time of Euplotes
was even lower than the lowest empirical velocity. This
indicates that there were factors affecting the settling
velocity, which cannot be modelled by our mathemat-
ical approach. The appearance of appendages, not
body shape, affected the settling velocity—a factor not
considered by Stokes law. Therefore, we had to reject
the first and the second hypotheses because sinking
velocities appear more strongly effected by the pre-
sence of appendages (cirri and membranelles) than
body shape. Hence, the experimental determination of

62
 M. CLAESSENS AND M. PRAST j SETTLING OF FIXED PLANKTON SAMPLES
sinking velocities is more reliable than the theoretical
approach.
The primary factors affecting the sinking velocities of
fixed ciliates were cell density and salinity of the habitat.
The marine ciliates had significantly lower settling vel-
ocities as the density of seawater is clearly higher than
the density of freshwater, whereas the density of the
marine ciliates was in the same range as that of the fresh-
water species, confirming the third hypothesis. The sal-
inity of the medium needs to be considered when settling
plankton samples, especially of ciliates, for enrichment.
Other factors affecting the sinking velocity in field
samples might be the presence of other organisms or exo-
polymeric substances. The cultures used for this study
were not axenic cultures, in addition to the ciliates they
also contained bacteria, flagellates and in case of the
marine assemblages, algae as well. Thus, the conditions in
the experiments reflected the conditions of field samples
from many marine and freshwater plankton habitats.
Furthermore, in comparison with field samples, our
experiments cover a wide range of salinity and size of
ciliates, but larger and smaller ciliates are common.
The calculation of the sinking velocities for S. epidemum
and P. tracheloides confirmed that size does not have a
major impact on sinking velocity. The resulting values
of 6.3 and 11.6 mm min21, respectively, were well
within the range of the calculated and experimentally
determined sinking velocities of the ciliate species used
in our experiments. However, the discrepancy between
the calculated and empirically determined sinking vel-
ocities and the differences in density between the ciliate
species shows that the experiments should be continued
with more ciliate species. This holds true especially for
tintinnid ciliates, which may carry a lorica that signifi-
cantly affects the density.
The sinking velocities measured and calculated in
this study exceeded the expectations by far. The result-
ing settling times are much shorter than the experience-
and estimation-derived times that are commonly in use
now. Compared to sinking velocities of other planktonic
particles like algae, the values of the fixed ciliate cells in
our study were very high. This might be due to the fact
that algae show an even wider morphological spectrum
(Padisák et al., 2003) and thus, possibly higher form
resistances, than ciliates, and that algae have various
features countering sedimentation, such as flotation
appendages and inclusions of lipids or gases. Many of
these keep their function after fixation, but this is not
true for ciliates, which use their cilia to prevent sinking.
Furthermore, algae often form colonies, which also
affect the sinking properties. For loricae of tintinnid cili-
ates, sinking velocities are between 0.1 and
11.4 mm min21 (Suzuki and Taniguchi, 1995).
63
S U M M A RY
The methods for determining ciliate density and settling
times both were established and can easily be repeated
for other species. As there were significant differences in
ciliate density between groups, this parameter should be
measured for each species on its own. The use of a
mean ciliate density cannot be recommended. For the
settling times, we recommend using the lowest empirical
sinking velocities, meaning 0.5 and 1.7 mm min21 for
fixed marine (salinities of 16 and 40, respectively) ciliate
Downloaded from https://academic.oup.com/plankt/article-abstract/30/1/57/1494425 by guest on 10 April 2019
samples and 2.4 mm min21 for fixed freshwater
samples. As a salinity of 40 is rather higher than typical
marine water, this value (1.7 mm min21) should be
reliable for most marine systems.
For Utermöhl settling chambers, the settling time can
be drastically decreased now, from at least 24 h for 10
and 50 mL samples to 1 h and 2.8 h for marine
samples and to the range of minutes for freshwater
samples (8.3 and 42 min). That means that 95% less
time is needed for settling ciliate samples, compared
with the old settling times.
AC K N OW L E D G E M E N T S
We thank W. Bauer (University Salzburg) for helping us
with the mathematic formulas and we are grateful to
S. Agatha and an anonymous reviewer for helpful com-
ments. The practical work was assisted by A. Breitner
and M. Vulelija.
FUNDING
Deutsche Forschungsgemeinschaft (DFG; WI 1623/3 to
S.A.Wickham) and by U.-G. Berninger and S.A.
Wickham.
REFERENCES
Bronstein, I. N. and Semendjajew, K. A. (1974) Taschenbuch der
Mathematik. BSB B. G. Teubner Verlagsgesellschaft.
Claessens, M., Wickham, S. A. and Post, A. F. High ciliate diversity
despite low resource levels in the oligotrophic Gulf of Aqaba, in
preperation.
Foissner, W. (1999) Identification and ecology of limnetic plankton
ciliates. Bayerisches Landesamt für Wasserwirtschaft Report Issue
3/99, Bartels & Wernitz, München.
Hamm, C. E. (2002) Interactive aggregation and sedimentation of
diatoms and clay-sized lithogenic material. Limnol. Oceanogr., 47,
1790–1795.
 JOURNAL OF PLANKTON RESEARCH j VOLUME
Kiorboe, T., Grossart, H.-P., Ploug, H. et al. (2002) Mechanisms and
rates of bacterial colonization of sinking aggregates. App. Envir.
Microbiol., 68, 3996–4006.
Kiorboe, T., Tang, K., Grossart, H.-P. et al. (2003) Dynamics of
microbial communities on marine snow aggregates: colonization,
growth, attachment, and grazing mortality of attached bacteria. App.
Envir. Microbiol., 69, 3036–3047.
Leakey, R. J. G., Burkill, P. H. and Sleigh, M. A. (1994) A comparison
of fixatives for the estimation of abundnace and biovolume of
marine planktonic ciliate populations. J. Plankton Res., 16, 375– 389.
McNown, J. S. and Malaika, J. (1950) Effects of particle shape on
settling velocity at low Reynolds numbers. Am. Geophys. Un., 31,
74– 81.
Padisák, J., Soróczki-Pintér, É. and Rezner, Z. (2003) Sinking proper-
ties of some phytoplankton shapes and the relation of form resist-
ance to morphological diversity of plankton—an experimental
study. Hydrobiologia, 500, 243– 257.
64
30 j NUMBER 1 j PAGES 57 – 64 j 2008
Peterson, M. L., Wakeham, S. G., Lee, C. et al. (2005) Novel tech-
niques for collection of sinking particles in the ocean and determin-
ing their settling rates. Limnol. Oceanogr.: Meth., 3, 520– 532.
Ptacnik, R., Diehl, S. and Berger, S. (2003) Performance of sinking
and nonsinking phytoplankton taxa in a gradient of mixing depth.
Limnol. Oceanogr., 48, 1903– 1912.
Stoecker, D. K., Gifford, D. J. and Putt, M. (1994) Preservation of
marine planktonic ciliates: losses and cell shrinkage during fixation.
Mar. Ecol. Prog. Ser., 110, 293–299.
Suzuki, T. and Taniguchi, A. (1995) Sinking rate of loricae of some
common tintinnid ciliates. Fish. Oceanogr., 4, 257– 263.
Downloaded from https://academic.oup.com/plankt/article-abstract/30/1/57/1494425 by guest on 10 April 2019
Utermöhl, H. (1958) Zur Vervollkommnung der quantitativen
Phytoplankton-Methodik. Mitt. Int. Verein. Limnol., 9, 1– 38.
Vogel, S. (1983) Life in Moving Fluids. Princeton University Press, pp.